{"id":"W4379511426","doi":"10.21428/594757db.6ee355e7","title":"Exploring Preferential Label Smoothing for Neural Network-based Classifiers","year":2023,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke; University of Alberta","funders":"","keywords":"Overfitting; Smoothing; Ground truth; Computer science; Regularization (linguistics); Artificial intelligence; Machine learning; Artificial neural network; Noise (video); Binary classification; Generalization; Pattern recognition (psychology); Mathematics; Support vector machine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004389247,0.0001038308,0.00009914281,0.0001036927,0.0002524866,0.0002305019,0.0005338942,0.00003549723,0.000009022968],"category_scores_gemma":[0.00009254497,0.00009327639,0.00004887256,0.0005096856,0.00001483664,0.0006185697,0.0001223995,0.0001312672,0.00007833893],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001922002,"about_ca_system_score_gemma":0.00003771254,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002574601,"about_ca_topic_score_gemma":0.00001323009,"domain_scores_codex":[0.9988714,0.000058192,0.0001709015,0.0003644559,0.0001800284,0.0003549664],"domain_scores_gemma":[0.9991566,0.0002318234,0.00006225096,0.0004332647,0.00004074888,0.00007530417],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004552096,0.00006357365,0.002600457,0.00008828349,0.00002889583,0.000003902528,0.0004551928,0.1104711,0.001633472,0.2365736,0.04920235,0.5988336],"study_design_scores_gemma":[0.0003837522,0.00004635096,0.005583778,0.000009800134,0.000003863694,3.535453e-7,0.00001665136,0.971724,0.0002003445,0.00053505,0.02137205,0.0001240377],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03488498,0.000008578719,0.9563822,0.004210822,0.001125908,0.0001988487,0.000003238865,0.001429339,0.001756158],"genre_scores_gemma":[0.9376735,0.000006879749,0.0601073,0.0004577507,0.0003385328,0.0001851038,0.0001231745,0.00001817115,0.001089598],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9027885,"threshold_uncertainty_score":0.3803701,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2684558467913105,"score_gpt":0.3194399407230387,"score_spread":0.0509840939317282,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}